Noometry

Model comparison

GPT-5.1 vs QwQ-32B

GPT-5.1 is the stronger model overall, scoring 48.8 to 39.4 on the Noometry Index.

Last verified . 29 shared benchmarks.

GPT-5.1 OpenAI

48.8

Rank #47 Confirmed

QwQ-32B Alibaba (Qwen)

39.4

Rank #153 Confirmed

Summary

  • They share 29 benchmarks with published results for both. GPT-5.1 scores higher in 7 categories and QwQ-32B in 1 category; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where GPT-5.1 leads 39.9 to 22.9.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 88.6% for GPT-5.1 and 59.2% for QwQ-32B.
  • QwQ-32B has downloadable open weights; the other is API-only.

Side by side

GPT-5.1QwQ-32B
ProviderOpenAIAlibaba (Qwen)
Noometry Index48.839.4
Released2025-11-132024-11-28
WeightsProprietaryOpen
Context window400K—
Max output128K—
Input $ / M tokens$1.25—
Output $ / M tokens$10—
Results tracked6336

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding GPT-5.1 leads

GPT-5.1: 46.9 (#59), QwQ-32B: 34.9 (#220)

BenchmarkGPT-5.1QwQ-32B
LiveBench Coding72.5%72.2%
LMArena Coding14541333
SWE-bench Verified68%—
SWE-bench Verified (bash only)66%—
Aider Polyglot—20.9%
LMArena WebDev1395—
SciCode43.3%—
GSO13.7%—
WeirdML60.8%—
BigCodeBench Instruct—44.6%
BigCodeBench Complete—54.4%
ALE-Bench1,192—

Agentic & Tool Use Not comparable

GPT-5.1: 32.5 (#55), QwQ-32B: —

BenchmarkGPT-5.1QwQ-32B
Terminal-Bench47.6%—
DeepResearch Bench42.8%—
LMArena Search1199—
Vending-Bench 21,473—

Reasoning GPT-5.1 leads

GPT-5.1: 39.9 (#50), QwQ-32B: 22.9 (#171)

BenchmarkGPT-5.1QwQ-32B
Chess Puzzles32%5%
LiveBench Reasoning95.8%83.5%
LMArena Hard Prompts14571326
LiveBench Data Analysis72.1%65%
Epoch Capabilities Index149.66137.6
ForecastBench58.158.3
LiveBench78.8%72%
ARC-AGI-217.6%—
SimpleBench53.2%—
ARC-AGI-172.8%—
CritPt4.9%—
EnigmaEval11.2%—
Mystery Game Puzzles19%—
DTBench90.1%—
LMCA43.9%—

Math GPT-5.1 leads

GPT-5.1: 50.3 (#45), QwQ-32B: 36.2 (#123)

BenchmarkGPT-5.1QwQ-32B
OTIS Mock AIME 2024-202588.6%59.2%
LiveBench Math94.5%77.8%
LMArena Math14461359
Omni-MATH46.4%—
FrontierMath (Feb 2025 set)31%—
FrontierMath Tier 4 (v1)12.5%—

Knowledge GPT-5.1 leads

GPT-5.1: 50.5 (#65), QwQ-32B: 37.2 (#150)

BenchmarkGPT-5.1QwQ-32B
GPQA Diamond87.6%65.3%
LMArena Expert14701323
Humanity's Last Exam23.7%—
SimpleQA Verified48%—
MMLU-Pro57.9%—
Confabulations—15.6%
Vectara Hallucination Rate10.9%—
GPQA (HELM)44.2%—

Multimodal Not comparable

GPT-5.1: 44.0 (#17), QwQ-32B: —

BenchmarkGPT-5.1QwQ-32B
LMArena Vision1250—
VPCT58.7%—
LMArena Document1403—

Multilingual GPT-5.1 leads

GPT-5.1: 54.0 (#51), QwQ-32B: 45.0 (#171)

BenchmarkGPT-5.1QwQ-32B
LMArena Non-English14311306
LMArena Chinese14951378
LMArena French14511337
LMArena German14391313
LMArena Japanese14531263
LMArena Korean14011278
LMArena Russian14351297
LMArena Spanish14301353

Instruction Following GPT-5.1 leads

GPT-5.1: 84.0 (#1), QwQ-32B: 72.8 (#131)

BenchmarkGPT-5.1QwQ-32B
LiveBench Instruction Following93.3%81.8%
LMArena Instruction Following14421297
IFEval93.5%—

Long Context QwQ-32B leads

GPT-5.1: 47.6 (#14), QwQ-32B: 49.0 (#11)

BenchmarkGPT-5.1QwQ-32B
LMArena Longer Query14481308
Fiction.LiveBench—83.3%
CL-bench23.7%—
CL-bench Life17.3%—

Writing & Preference GPT-5.1 leads

GPT-5.1: 64.7 (#53), QwQ-32B: 50.9 (#175)

BenchmarkGPT-5.1QwQ-32B
LMArena Text14431329
LMArena Creative Writing14271288
LMArena Multi-Turn14491314
LiveBench Language80.2%51.4%
Short-Story Creative Writing—80.2%
EQ-Bench Creative Writing—1257
WildBench86.3%—

Frequently asked questions

Is GPT-5.1 better than QwQ-32B?

GPT-5.1 is the stronger model overall, scoring 48.8 to 39.4 on the Noometry Index.

Is GPT-5.1 or QwQ-32B better for coding?

GPT-5.1 scores higher on coding benchmarks: 46.9 versus 34.9 in the Noometry coding category.

How many benchmarks do GPT-5.1 and QwQ-32B share?

29 benchmarks have published results for both models. GPT-5.1 has 63 scored results on Noometry and QwQ-32B has 36.

Related comparisons

Go deeper